Toolik - summer 2019 field campaign
by Hélène Angot
Gas tanks
Lag time fast systems
Surface O3 concentrations
TO DO
- Fast NOx: puff system.
- WS/WD sensor - compare WS/WD data from tower with station.
- Write code to compare clocks on fast O3 and fast NOx laptops.
Thursday, May 23rd
Daily checks
- Lab temperature = 20.7°C, AC temperature = 68F.
- Case temperature: fast O3 = ~22.2°C, fast NOx = ~18.8°C.
- Drierite OK.
- Fast O3: OK, NO flow rate OK.
- Fast NOx: OK.
Puff system
5 injections at (fast O3 time):
2019-05-23 08:03:33
2019-05-23 08:11:20
2019-05-23 08:12:30
2019-05-23 08:13:59
2019-05-23 08:15:22
GC/MS
On VI inlet system “Indicators”:
* Advance GC start (s) = 10 instead of 360.
* GC run time (s) = 1200 instead of 1500.
NPL run (600 s): TL_20190521_13
NPL run is good. New sequence started: GCMS_Toolik_20190523.sequence.
Inlet system:
0) Std2 BVOC
2) to 13) Ambient BVOC
14) Reset.
Time between samples = 4200 s.
Wednesday, May 22nd
Daily checks
- Lab temperature = 21.7°C, AC temperature = 68F.
- Case temperature: fast O3 = ~22.7°C, fast NOx = ~18.6°C.
- Drierite OK.
- Fast O3: OK, NO flow rate OK.
- Fast NOx: OK.
- GC/MS: running, water level OK.
- Gas tanks: OK.
Fast NOx/O3 systems
- Reorganize sample line
The sample line first goes to the fast NOx, then to the fast O3. However, while the fast NOx is pumping 1 L/min, we send 1.2 L/min of Zero Air during calibrations. 0.2 L/min doesn’t make it to the instrument and goes back in the sample line, to the fast O3. During a calibration, a fraction of NO can thus be transported through the sample line to the fast O3 (and this NO will react with O3). Jacques suggested that we switch the order of the “tees” connected to the sample line so that the fast O3 is connected first.
Done 9:40-9:53am AST.
- Puff system
I performed several injections between 10:40-11:00am AST. Unfortunately, the status of the fast NOx lag time valve is not saved in the VI (Jacques saved the fast O3 lag time valve instead). He will update the VI and I’ll do more puff tests when it’s done.
GC/MS
The chromatograms look weird, at least very different from what we got last year.
Today’s chromatogram:
Last year:
I compared the acquisition methods (in :D/MassHunter/GCMS/1/data/Toolik_GCMS_data/acqmeth.text). While Front SSZ Inlet He used to be at 200°C, Detlev changed it to 100°C on 12/09/2018. According to Jacques, this shouldn’t make any difference.
I connected the NPL tank to STD3 inlet.
Inlet system sequence:
0) STD3 BVOC (2400 s)
1) Stop
This is chromatogram TL_20190521_12.
NPL run last summmer (TL_20180728_08):
The separation of the peaks is not as good. Jacques remembers that there is now “Advanced GC start = 360 s” on the inlet system VI (Detlev asked for this), meaning that the GC actually starts 6 min earlier - the oven temperature is thus higher (because it’s been running for 6 more minutes) when the injection is done. This influences the retention times and potentially the separation.
Miscellaneous
- We synchronized the fast O3 and the PTR-TOF-MS clocks.
Control Panel > Date and Time > Internet Time > Check Automatically synchronize with an internet time server: time.nist.gov. Synchronization done at 3:04 pm AST. The fast O3 is now ~1min30s ahead of fast NOx.
- Anssi built the tent! The one we bought is too small (6 x 6 x 6) for our two autosamplers and Anssi’s instrument. He bought the same model but 8 x 8 x 8 instead - we’re using his tent.
Tuesday, May 21st
Fast O3 calibration
Based on the calibration performed on Sunday, we know that the fast O3 is biased low by ~28-30%. Conversion from counts to ppb has to be updated.
O3 span = 0.000445 becomes 0.000612
offset = 0.11 becomes -1.47
Changes made to VI at 6:46pm AST.
Data from the slow and fast O3 are now ~comparable (keeping in mind that slow O3 was 3% too low compared to calibrator).
WS/WD sensor
Our sensor is off by ~80 degrees compared to data from the station (data available here.). I adjusted the position of the sensor accordingly at ~4pm local time.
Drierite
Drierite changed on both fast NOx and O3 (at ~1:50-2pm local time). Put drierite in oven at 210°C for 1 hr (LAB 2).
GC/MS
The settings in the sequence of the inlet system were wrong; use settings from last summer (BVOC settings are different from OVOC settings):
* sample time = 2400
* sample flow = 1.25
* backflush time = 180
* desorb time = 300
* flash heat time = 2
* inject time = 30
* bakeout temp = 325 * bakeout time = 40
* carrier flow = 1.25
Started new sequence on GC/MS (and inlet system): GCMS_Toolik_20100521.sequence.xml
Fast NOx
Note that the fast NOx data are now being saved in the fast O3 data files.
Columns headers:
* date fast O3 * time fast O3 (hhmm)
* time fast O3 (ss.s)
* counts
* O3 concentration (ppb)
* 0 fast O3 - NO valve status or lag time valve
* 1 fast O3 - NO valve status or lag time valve
* date fast NOx
* time fast NOx (hhmm)
* time fast NOx (ss.s)
* counts
* 0 or 1 (1 = auto mode)
* 0 or 1 (if NO or NOx)
* 0 or 1 (1 if calibration)
* 0 to 7: step in calibration
I performed a few tests with Jacques.
* O2 generator not stable. Changed the fittings at the outlet (see pic below), was damaged. O2 flow is now stable. Jacques still wonders whether the O2 might be contaminated (we see spikes from time to time) - Order ultra high purity O2.
- The signal was still unstable and the background occasionally pretty high. We performed several tests in order to identify a potential leak. Couldn’t find anything. We ended up disconnecting/reconnecting the PMT high voltage and signal cables (at the top rear of the instrument - see pic below).
Monday, May 20th
- Tee on Zero Air line for PTR-TOF-MS.
WS/WD sensor
I need to replace the Reed Switch.
1. Disconnect WS/WD rear of MET Controls box
2. Disconnect anemometer from tower
According to instructions:
3. Open the sensor housing
4. Disconnect the wires
TOP to BOTTOM: RED, BLACK, YELLOW, GREEN, WHITE.
5. Push switch cartridge out of its seat. Clean off any old sealant from the top plate assembly.
Done
The WS constantly increases. Something’s wrong. Jacques suggests that the sensor counts the # of rotations but doesn’t seem to convert to m/s. Something’s wrong in the VI. Problem fixed! Jacques made changes in the VI.
New VI: METTower_Toolik_190520.vi
Daily checks
- Lab temperature = 21°C, AC temperature = 68F.
- Case temperature: fast O3 = ~22°C, fast NOx = ~17.6°C.
- Drierite OK.
- Fast O3: OK, NO flow rate OK.
- Fast NOx: OK.
- GC/MS: running, water level OK.
- Gas tanks: OK.
Inventory
- I talked to Justin (Toolik manager) about the canopy tent. We agreed that we can build it without asking BLM (quote: “better ask for forgiveness than permission”). He will let us know if BLM comes visit - if so, we will pack everything. The plan is for Anssi and I to build it tomorrow. It’s going to rain Wed/Thur/Fri so we can make sure it’s waterproof before we install instruments in it.
- O3 sonde prep unit + O3 sondes (x2) + NOAA laptop: LAB1.
- Helium tanks for balloon (balloon grade): 3 large ones + 1 smaller one.
- Kestrel in Box#2 (warm storage).
- Kestrel works with 1 AA battery.
- Cold storage (south entrance): CPS is currently using the building for construction work/storage. They’re leaving Toolik on June 6th. I’ll check with them if we can use the building before that to store the balloon.
